problem with view of list collection in history view
Hi Team, my tool creates dynamically 96 datasets bundled into a list. In the history I can see the number 96 in the top as hidden datasets (6 shown, 96 hidden) When I open the list, I just can see 64 items. Now I run the job again and I have 96 more hidden items. I open the new list and can see 66 items in that new list. What is going on here? Is that "just" a visual bug? Or are my datasets affected? Thanks, Alexander PS: I use postgres
Conversation in IRC. tl;dr - it looks like it might be a GUI related problem since the API does contain all of the datasets. Carl - any chance you have an idea of what is going on here? 21:20 < jmchilton> avowinkel: is it possible there were duplicated identifiers (has your discover_datasets pattern changed from earlier) 21:21 < jmchilton> I'm leaning toward saying it is likely a backend problem - since explicit output collections are pretty new and you are the first person I can think of really exercising them strenuously 21:21 < jmchilton> One way to verify though is to check the API - if you just open localhost:<port>/api/histories in your browser - find the history id 21:21 < jmchilton> then open /api/histories/<history_id>/contents and then find the collection 21:22 < jmchilton> you should be able to open something like 21:22 < jmchilton> /api/histories/<history_id>/contents/collections/<collection_id> - which should show the individual datasets 21:23 < avowinkel> there are defenitely no duplicate designations - if thats the same like identifiers 21:23 < avowinkel> It's still <discover_datasets pattern="__name_and_ext__" directory="splits" /> 21:27 < jmchilton> My next question would be (if you can verify it is a backend thing) - are the elements in the dataset - the hidden elements less than a certain HID - or are they random. 21:28 < avowinkel> via the api all 96 entries are in the collection 21:29 < avowinkel> with element_index's 0 to 95, in total 96 21:29 < avowinkel> in both lists 21:31 < avowinkel> biggest hid is 202 21:32 < avowinkel> the parent's list hid is always smaller than the containing element's hids 22:26 < jmchilton> so you are sure every element_index from 0 to 95 is represented? This being a GUI problem is really odd - but it seems like it probably is. I wonder if someone a div id is generated from the identifiers in such a way that one is duplicated. Seems unlikely 22:26 < jmchilton> Can you open your JavaScript console and see if there are any JavaScript errors/ 22:27 < avowinkel> well. I did grep element_index, I saw index 0 on the top, Index 95 on the bottom. and wc -l gives 96 - so yes. very sure 22:28 < avowinkel> and when I scan loosely through the list of greps, I don't see anything odd 22:29 < avowinkel> don't want to count from 0 to 95 ^^ 22:29 < jmchilton> :) 22:29 < avowinkel> for all the tests 22:30 < jmchilton> does that API response have a hidden field for the datasets? 22:31 < avowinkel> there is nothing in that file that matches "hidden" 22:31 < avowinkel> (in the history they are all hidden) 22:32 < jmchilton> I would open your web browser and check for javascript errors next 22:34 < avowinkel> nop. nothing (Firefox 34 - ubuntu biolinux) 22:34 < jmchilton> can you send me a screenshot of the expanded collection? 22:35 < avowinkel> the newest run has 69 entries in the history 22:36 < avowinkel> what part do you want screenshotted? 22:38 < jmchilton> "When I open the list, I just can see 64 items." The opened list in the history panel 22:40 < avowinkel> http://snag.gy/2knoI.jpg 22:43 < jmchilton> are you hand counting these lists in the browswer then? 22:47 < avowinkel> yes, hand counting 22:50 < jmchilton> I'll ping carl about this - he is the GUI mastermind - he might have some clue 22:59 < avowinkel> jmchilton: http://pastebin.com/DcpF1QAU 22:59 < mrscribe> Title: [YAML] galaxy dataset_collection contents - Pastebin.com (at pastebin.com) 23:00 < avowinkel> don't get confused: On the picture is a different dataset. It doesn't have "sample_" in the name 23:04 < jmchilton> yeah - that response looks perfectly fine - really odd -John On Thu, Jun 25, 2015 at 4:58 PM, Alexander Vowinkel <vowinkel.alexander@gmail.com> wrote:
Hi Team,
my tool creates dynamically 96 datasets bundled into a list. In the history I can see the number 96 in the top as hidden datasets (6 shown, 96 hidden)
When I open the list, I just can see 64 items.
Now I run the job again and I have 96 more hidden items. I open the new list and can see 66 items in that new list.
What is going on here? Is that "just" a visual bug? Or are my datasets affected?
Thanks, Alexander
PS: I use postgres
___________________________________________________________ Please keep all replies on the list by using "reply all" in your mail client. To manage your subscriptions to this and other Galaxy lists, please use the interface at: https://lists.galaxyproject.org/
To search Galaxy mailing lists use the unified search at: http://galaxyproject.org/search/mailinglists/
Thanks for the report Alexander. Carl has fixed this problem in dev with https://github.com/galaxyproject/galaxy/pull/431. Thanks Carl! -John On Thu, Jun 25, 2015 at 7:11 PM, John Chilton <jmchilton@gmail.com> wrote:
Conversation in IRC. tl;dr - it looks like it might be a GUI related problem since the API does contain all of the datasets. Carl - any chance you have an idea of what is going on here?
21:20 < jmchilton> avowinkel: is it possible there were duplicated identifiers (has your discover_datasets pattern changed from earlier) 21:21 < jmchilton> I'm leaning toward saying it is likely a backend problem - since explicit output collections are pretty new and you are the first person I can think of really exercising them strenuously 21:21 < jmchilton> One way to verify though is to check the API - if you just open localhost:<port>/api/histories in your browser - find the history id 21:21 < jmchilton> then open /api/histories/<history_id>/contents and then find the collection 21:22 < jmchilton> you should be able to open something like 21:22 < jmchilton> /api/histories/<history_id>/contents/collections/<collection_id> - which should show the individual datasets 21:23 < avowinkel> there are defenitely no duplicate designations - if thats the same like identifiers 21:23 < avowinkel> It's still <discover_datasets pattern="__name_and_ext__" directory="splits" /> 21:27 < jmchilton> My next question would be (if you can verify it is a backend thing) - are the elements in the dataset - the hidden elements less than a certain HID - or are they random. 21:28 < avowinkel> via the api all 96 entries are in the collection 21:29 < avowinkel> with element_index's 0 to 95, in total 96 21:29 < avowinkel> in both lists 21:31 < avowinkel> biggest hid is 202 21:32 < avowinkel> the parent's list hid is always smaller than the containing element's hids 22:26 < jmchilton> so you are sure every element_index from 0 to 95 is represented? This being a GUI problem is really odd - but it seems like it probably is. I wonder if someone a div id is generated from the identifiers in such a way that one is duplicated. Seems unlikely 22:26 < jmchilton> Can you open your JavaScript console and see if there are any JavaScript errors/ 22:27 < avowinkel> well. I did grep element_index, I saw index 0 on the top, Index 95 on the bottom. and wc -l gives 96 - so yes. very sure 22:28 < avowinkel> and when I scan loosely through the list of greps, I don't see anything odd 22:29 < avowinkel> don't want to count from 0 to 95 ^^ 22:29 < jmchilton> :) 22:29 < avowinkel> for all the tests 22:30 < jmchilton> does that API response have a hidden field for the datasets? 22:31 < avowinkel> there is nothing in that file that matches "hidden" 22:31 < avowinkel> (in the history they are all hidden) 22:32 < jmchilton> I would open your web browser and check for javascript errors next 22:34 < avowinkel> nop. nothing (Firefox 34 - ubuntu biolinux) 22:34 < jmchilton> can you send me a screenshot of the expanded collection? 22:35 < avowinkel> the newest run has 69 entries in the history 22:36 < avowinkel> what part do you want screenshotted? 22:38 < jmchilton> "When I open the list, I just can see 64 items." The opened list in the history panel 22:40 < avowinkel> http://snag.gy/2knoI.jpg 22:43 < jmchilton> are you hand counting these lists in the browswer then? 22:47 < avowinkel> yes, hand counting 22:50 < jmchilton> I'll ping carl about this - he is the GUI mastermind - he might have some clue 22:59 < avowinkel> jmchilton: http://pastebin.com/DcpF1QAU 22:59 < mrscribe> Title: [YAML] galaxy dataset_collection contents - Pastebin.com (at pastebin.com) 23:00 < avowinkel> don't get confused: On the picture is a different dataset. It doesn't have "sample_" in the name 23:04 < jmchilton> yeah - that response looks perfectly fine - really odd
-John
On Thu, Jun 25, 2015 at 4:58 PM, Alexander Vowinkel <vowinkel.alexander@gmail.com> wrote:
Hi Team,
my tool creates dynamically 96 datasets bundled into a list. In the history I can see the number 96 in the top as hidden datasets (6 shown, 96 hidden)
When I open the list, I just can see 64 items.
Now I run the job again and I have 96 more hidden items. I open the new list and can see 66 items in that new list.
What is going on here? Is that "just" a visual bug? Or are my datasets affected?
Thanks, Alexander
PS: I use postgres
___________________________________________________________ Please keep all replies on the list by using "reply all" in your mail client. To manage your subscriptions to this and other Galaxy lists, please use the interface at: https://lists.galaxyproject.org/
To search Galaxy mailing lists use the unified search at: http://galaxyproject.org/search/mailinglists/
participants (2)
-
Alexander Vowinkel -
John Chilton